© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38881
    Prev
    Next
    INDEX
    Explanations

    place names or proper nouns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    èĮ¸
    -0.11
    èĽŁ
    -0.10
    ä¸Ģ对ä¸Ģ
    -0.10
    çŀĢ
    -0.09
    éĻ¢éķ¿
    -0.09
    è´²
    -0.09
    æ²®
    -0.09
    æī¯
    -0.09
    ä¼ļéķ¿
    -0.09
    ibil
    -0.09
    POSITIVE LOGITS
     sectional
    0.08
    æľīå¤ļå°ij
    0.08
    å°ijãģªãģı
    0.08
    iant
    0.08
    0
    0.08
    acon
    0.08
    lim
    0.08
    7
    0.08
    bench
    0.08
    9
    0.08
    Activations Density 0.156%

    No Known Activations